Impact of clinical case discussions in anatomy on national exam performance: a 10-year longitudinal cohort study

Traditional anatomy instruction often prioritizes structural memorization over clinical application. Although case-based learning is widely endorsed to bridge this theory-practice gap, longitudinal evidence regarding its sustained impact in foundational sciences remains limited, particularly concerning its performance over a decade amid massive enrollment expansion.

This 10-year longitudinal retrospective cohort study tracked seven clinical medicine cohorts (2015-2021). To maintain teaching quality amid a massive enrollment expansion initiated in 2018, cohorts 2015-2017 received traditional instruction (control), while 2018-2021 underwent structured clinical case discussions (experimental). Primary outcomes were anatomy mastery rate differentials (university performance minus national benchmarks) from two national examinations: the Proficiency Test (stratified into integrated and non-integrated questions) and the National Medical Licensing Examination (overall mastery).

In the Proficiency Test, the experimental group outperformed the national average in integrated questions, peaking at +3.37 (2018 cohort), while non-integrated differentials reached +5.91 (2020 cohort). However, the 2021 cohort's integrated differential sharply declined to -5.34, coinciding temporally with the stagnation of the case repository (the case scenarios had remained unchanged since the 2020 cohort). In the National Medical Licensing Examination, the experimental group's overall differential peaked at +5.69 (2019 cohort). Despite enrollment expanding from 408 to a peak of 795 students, performance remained stable across both examinations.

The implementation of clinical case discussions was associated with improved performance on integrated clinical-anatomy questions in national examinations, and performance remained stable despite substantial enrollment expansion. The observed decline in the 2021 cohort coincided with stagnation of the case repository, suggesting that periodic content renewal may be important for sustaining pedagogical effectiveness. These findings highlight the need for future research on dynamic case repositories and adaptive assessment strategies in the context of emerging educational technologies.